Limits...
Bacteria dialog with Santa Rosalia: Are aggregations of cosmopolitan bacteria mainly explained by habitat filtering or by ecological interactions?

Pascual-García A, Tamames J, Bastolla U - BMC Microbiol. (2014)

Bottom Line: Due to the coarse grained nature of our data we cannot conclusively reject the hypothesis that many of these aggregations are due to environmental preferences that the model fails to reproduce.As observed in similar studies, we also found that phylogenetically related taxa are much more prone to aggregate than to segregate, an observation that may play a role in bacterial speciation.We hope that these results stimulate experimental verification of the putative cooperative interactions between cosmopolitan bacteria, and we suggest several groups of aggregated cosmopolitan bacteria that are interesting candidates for such an investigation.

View Article: PubMed Central - PubMed

Affiliation: Centro de Biología Molecular Severo Ochoa (CSIC-UAM), c. Nicolás Cabrera 1, campus UAM, Madrid, E-28049, Spain. alberto.pascual@csic.es.

ABSTRACT

Background: Since the landmark Santa Rosalia paper by Hutchinson, niche theory addresses the determinants of biodiversity in terms of both environmental and biological aspects. Disentangling the role of habitat filtering and interactions with other species is critical for understanding microbial ecology. Macroscopic biogeography explores hypothetical ecological interactions through the analysis of species associations. These methods have started to be incorporated into microbial ecology relatively recently, due to the inherent experimental difficulties and the coarse grained nature of the data.

Results: Here we investigate the influence of environmental preferences and ecological interactions in the tendency of bacterial taxa to either aggregate or segregate, using a comprehensive dataset of bacterial taxa observed in a wide variety of environments. We assess significance of taxa associations through a model that takes into account habitat preferences and the global distribution of taxa across samples. The analysis of these associations reveals a surprisingly large number of significant aggregations between taxa, with a marked community structure and a strong propensity to aggregate for cosmopolitan taxa. Due to the coarse grained nature of our data we cannot conclusively reject the hypothesis that many of these aggregations are due to environmental preferences that the model fails to reproduce. Nevertheless, some observations are better explained by ecological interactions than by habitat filtering. In particular, most pairs of aggregating taxa co-occur in very different environments, which makes it unlikely that these associations are due to habitat preferences, and many are formed by cosmopolitan taxa without well defined habitat preferences. Moreover, known cooperative interactions are retrieved as aggregating pairs of taxa. As observed in similar studies, we also found that phylogenetically related taxa are much more prone to aggregate than to segregate, an observation that may play a role in bacterial speciation.

Conclusions: We hope that these results stimulate experimental verification of the putative cooperative interactions between cosmopolitan bacteria, and we suggest several groups of aggregated cosmopolitan bacteria that are interesting candidates for such an investigation.

Show MeSH
Number of aggregations of bacterial taxa divided by number of samples in which the taxon is present versus community cosmopolitanism Eq.(1). The black lines are obtained for the observed network, the red lines are obtained for a control network. Each point represents the average of all taxa in the same bin of cosmopolitanism.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4263022&req=5

Fig4: Number of aggregations of bacterial taxa divided by number of samples in which the taxon is present versus community cosmopolitanism Eq.(1). The black lines are obtained for the observed network, the red lines are obtained for a control network. Each point represents the average of all taxa in the same bin of cosmopolitanism.

Mentions: The number of aggregations of a taxon is positively correlated with its cosmopolitanism both for the control and for the observed network, but in the latter case the correlation is much stronger (r=0.64 instead of r=0.35). If we normalize the number of aggregations dividing it by the number of samples in which the taxon is present, called prevalence, the relationship with cosmopolitanism remains positive for the observed network whereas it becomes negative for the control network (see Figure 4 for community cosmopolitanism and Additional file 1: Figure S4 for environmental cosmopolitanism). This qualitative difference suggests that the observed relation between aggregations and cosmopolitanism goes beyond the trivial effect that more common taxa are more likely to co-occur. Since cosmopolitan taxa do not present well-defined preferences, it seems unlikely that the excess aggregation is due to habitat filtering. For instance, Flavobacterium and Pseudomonas are present in 36 different subtypes such as Arctic, Mouse Gut, Food Treatment or Mines among others. The hypothesis that their cooccurrence is explained by habitat preferences would imply that these hypothetical preferred properties co-occur in such a wide variety of environments. A more economical hypothesis is that the excess of co-occurrence is explained by cooperative interactions. Another possible hypothesis is that two cosmopolitan have an indirect relationship, due to the fact that there are specialist taxa that, if present, exclude both of them. Our data do not allow distinguishing between direct and indirect relationships, therefore we cannot judge how likely is this hypothesis.Figure 4


Bacteria dialog with Santa Rosalia: Are aggregations of cosmopolitan bacteria mainly explained by habitat filtering or by ecological interactions?

Pascual-García A, Tamames J, Bastolla U - BMC Microbiol. (2014)

Number of aggregations of bacterial taxa divided by number of samples in which the taxon is present versus community cosmopolitanism Eq.(1). The black lines are obtained for the observed network, the red lines are obtained for a control network. Each point represents the average of all taxa in the same bin of cosmopolitanism.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4263022&req=5

Fig4: Number of aggregations of bacterial taxa divided by number of samples in which the taxon is present versus community cosmopolitanism Eq.(1). The black lines are obtained for the observed network, the red lines are obtained for a control network. Each point represents the average of all taxa in the same bin of cosmopolitanism.
Mentions: The number of aggregations of a taxon is positively correlated with its cosmopolitanism both for the control and for the observed network, but in the latter case the correlation is much stronger (r=0.64 instead of r=0.35). If we normalize the number of aggregations dividing it by the number of samples in which the taxon is present, called prevalence, the relationship with cosmopolitanism remains positive for the observed network whereas it becomes negative for the control network (see Figure 4 for community cosmopolitanism and Additional file 1: Figure S4 for environmental cosmopolitanism). This qualitative difference suggests that the observed relation between aggregations and cosmopolitanism goes beyond the trivial effect that more common taxa are more likely to co-occur. Since cosmopolitan taxa do not present well-defined preferences, it seems unlikely that the excess aggregation is due to habitat filtering. For instance, Flavobacterium and Pseudomonas are present in 36 different subtypes such as Arctic, Mouse Gut, Food Treatment or Mines among others. The hypothesis that their cooccurrence is explained by habitat preferences would imply that these hypothetical preferred properties co-occur in such a wide variety of environments. A more economical hypothesis is that the excess of co-occurrence is explained by cooperative interactions. Another possible hypothesis is that two cosmopolitan have an indirect relationship, due to the fact that there are specialist taxa that, if present, exclude both of them. Our data do not allow distinguishing between direct and indirect relationships, therefore we cannot judge how likely is this hypothesis.Figure 4

Bottom Line: Due to the coarse grained nature of our data we cannot conclusively reject the hypothesis that many of these aggregations are due to environmental preferences that the model fails to reproduce.As observed in similar studies, we also found that phylogenetically related taxa are much more prone to aggregate than to segregate, an observation that may play a role in bacterial speciation.We hope that these results stimulate experimental verification of the putative cooperative interactions between cosmopolitan bacteria, and we suggest several groups of aggregated cosmopolitan bacteria that are interesting candidates for such an investigation.

View Article: PubMed Central - PubMed

Affiliation: Centro de Biología Molecular Severo Ochoa (CSIC-UAM), c. Nicolás Cabrera 1, campus UAM, Madrid, E-28049, Spain. alberto.pascual@csic.es.

ABSTRACT

Background: Since the landmark Santa Rosalia paper by Hutchinson, niche theory addresses the determinants of biodiversity in terms of both environmental and biological aspects. Disentangling the role of habitat filtering and interactions with other species is critical for understanding microbial ecology. Macroscopic biogeography explores hypothetical ecological interactions through the analysis of species associations. These methods have started to be incorporated into microbial ecology relatively recently, due to the inherent experimental difficulties and the coarse grained nature of the data.

Results: Here we investigate the influence of environmental preferences and ecological interactions in the tendency of bacterial taxa to either aggregate or segregate, using a comprehensive dataset of bacterial taxa observed in a wide variety of environments. We assess significance of taxa associations through a model that takes into account habitat preferences and the global distribution of taxa across samples. The analysis of these associations reveals a surprisingly large number of significant aggregations between taxa, with a marked community structure and a strong propensity to aggregate for cosmopolitan taxa. Due to the coarse grained nature of our data we cannot conclusively reject the hypothesis that many of these aggregations are due to environmental preferences that the model fails to reproduce. Nevertheless, some observations are better explained by ecological interactions than by habitat filtering. In particular, most pairs of aggregating taxa co-occur in very different environments, which makes it unlikely that these associations are due to habitat preferences, and many are formed by cosmopolitan taxa without well defined habitat preferences. Moreover, known cooperative interactions are retrieved as aggregating pairs of taxa. As observed in similar studies, we also found that phylogenetically related taxa are much more prone to aggregate than to segregate, an observation that may play a role in bacterial speciation.

Conclusions: We hope that these results stimulate experimental verification of the putative cooperative interactions between cosmopolitan bacteria, and we suggest several groups of aggregated cosmopolitan bacteria that are interesting candidates for such an investigation.

Show MeSH